forked from phoenix/litellm-mirror
use async functions
This commit is contained in:
parent
16c0307eab
commit
d71b86cd24
2 changed files with 30 additions and 10 deletions
|
@ -2454,6 +2454,17 @@
|
||||||
"mode": "chat",
|
"mode": "chat",
|
||||||
"source": "https://cloud.google.com/vertex-ai/generative-ai/pricing#partner-models"
|
"source": "https://cloud.google.com/vertex-ai/generative-ai/pricing#partner-models"
|
||||||
},
|
},
|
||||||
|
"vertex_ai/meta/llama-3.2-90b-vision-instruct-maas": {
|
||||||
|
"max_tokens": 8192,
|
||||||
|
"max_input_tokens": 128000,
|
||||||
|
"max_output_tokens": 8192,
|
||||||
|
"input_cost_per_token": 0.0,
|
||||||
|
"output_cost_per_token": 0.0,
|
||||||
|
"litellm_provider": "vertex_ai-llama_models",
|
||||||
|
"mode": "chat",
|
||||||
|
"supports_system_messages": true,
|
||||||
|
"source": "https://console.cloud.google.com/vertex-ai/publishers/meta/model-garden/llama-3.2-90b-vision-instruct-maas"
|
||||||
|
},
|
||||||
"vertex_ai/mistral-large@latest": {
|
"vertex_ai/mistral-large@latest": {
|
||||||
"max_tokens": 8191,
|
"max_tokens": 8191,
|
||||||
"max_input_tokens": 128000,
|
"max_input_tokens": 128000,
|
||||||
|
|
|
@ -901,9 +901,6 @@ async def update_database(
|
||||||
- Update litellm-proxy-budget row (global proxy spend)
|
- Update litellm-proxy-budget row (global proxy spend)
|
||||||
"""
|
"""
|
||||||
## if an end-user is passed in, do an upsert - we can't guarantee they already exist in db
|
## if an end-user is passed in, do an upsert - we can't guarantee they already exist in db
|
||||||
existing_token_obj = await user_api_key_cache.async_get_cache(
|
|
||||||
key=hashed_token
|
|
||||||
)
|
|
||||||
existing_user_obj = await user_api_key_cache.async_get_cache(key=user_id)
|
existing_user_obj = await user_api_key_cache.async_get_cache(key=user_id)
|
||||||
if existing_user_obj is not None and isinstance(existing_user_obj, dict):
|
if existing_user_obj is not None and isinstance(existing_user_obj, dict):
|
||||||
existing_user_obj = LiteLLM_UserTable(**existing_user_obj)
|
existing_user_obj = LiteLLM_UserTable(**existing_user_obj)
|
||||||
|
@ -1156,7 +1153,9 @@ async def update_cache(
|
||||||
|
|
||||||
# Update the cost column for the given token
|
# Update the cost column for the given token
|
||||||
existing_spend_obj.spend = new_spend
|
existing_spend_obj.spend = new_spend
|
||||||
user_api_key_cache.set_cache(key=hashed_token, value=existing_spend_obj)
|
await user_api_key_cache.async_set_cache(
|
||||||
|
key=hashed_token, value=existing_spend_obj
|
||||||
|
)
|
||||||
|
|
||||||
### UPDATE USER SPEND ###
|
### UPDATE USER SPEND ###
|
||||||
async def _update_user_cache():
|
async def _update_user_cache():
|
||||||
|
@ -1185,10 +1184,12 @@ async def update_cache(
|
||||||
# Update the cost column for the given user
|
# Update the cost column for the given user
|
||||||
if isinstance(existing_spend_obj, dict):
|
if isinstance(existing_spend_obj, dict):
|
||||||
existing_spend_obj["spend"] = new_spend
|
existing_spend_obj["spend"] = new_spend
|
||||||
user_api_key_cache.set_cache(key=_id, value=existing_spend_obj)
|
await user_api_key_cache.async_set_cache(
|
||||||
|
key=_id, value=existing_spend_obj
|
||||||
|
)
|
||||||
else:
|
else:
|
||||||
existing_spend_obj.spend = new_spend
|
existing_spend_obj.spend = new_spend
|
||||||
user_api_key_cache.set_cache(
|
await user_api_key_cache.async_set_cache(
|
||||||
key=_id, value=existing_spend_obj.json()
|
key=_id, value=existing_spend_obj.json()
|
||||||
)
|
)
|
||||||
## UPDATE GLOBAL PROXY ##
|
## UPDATE GLOBAL PROXY ##
|
||||||
|
@ -1237,10 +1238,14 @@ async def update_cache(
|
||||||
# Update the cost column for the given user
|
# Update the cost column for the given user
|
||||||
if isinstance(existing_spend_obj, dict):
|
if isinstance(existing_spend_obj, dict):
|
||||||
existing_spend_obj["spend"] = new_spend
|
existing_spend_obj["spend"] = new_spend
|
||||||
user_api_key_cache.set_cache(key=_id, value=existing_spend_obj)
|
await user_api_key_cache.async_set_cache(
|
||||||
|
key=_id, value=existing_spend_obj
|
||||||
|
)
|
||||||
else:
|
else:
|
||||||
existing_spend_obj.spend = new_spend
|
existing_spend_obj.spend = new_spend
|
||||||
user_api_key_cache.set_cache(key=_id, value=existing_spend_obj.json())
|
await user_api_key_cache.async_set_cache(
|
||||||
|
key=_id, value=existing_spend_obj.json()
|
||||||
|
)
|
||||||
except Exception as e:
|
except Exception as e:
|
||||||
verbose_proxy_logger.exception(
|
verbose_proxy_logger.exception(
|
||||||
f"An error occurred updating end user cache: {str(e)}"
|
f"An error occurred updating end user cache: {str(e)}"
|
||||||
|
@ -1279,10 +1284,14 @@ async def update_cache(
|
||||||
# Update the cost column for the given user
|
# Update the cost column for the given user
|
||||||
if isinstance(existing_spend_obj, dict):
|
if isinstance(existing_spend_obj, dict):
|
||||||
existing_spend_obj["spend"] = new_spend
|
existing_spend_obj["spend"] = new_spend
|
||||||
user_api_key_cache.set_cache(key=_id, value=existing_spend_obj)
|
await user_api_key_cache.async_set_cache(
|
||||||
|
key=_id, value=existing_spend_obj
|
||||||
|
)
|
||||||
else:
|
else:
|
||||||
existing_spend_obj.spend = new_spend
|
existing_spend_obj.spend = new_spend
|
||||||
user_api_key_cache.set_cache(key=_id, value=existing_spend_obj)
|
await user_api_key_cache.async_set_cache(
|
||||||
|
key=_id, value=existing_spend_obj
|
||||||
|
)
|
||||||
except Exception as e:
|
except Exception as e:
|
||||||
verbose_proxy_logger.exception(
|
verbose_proxy_logger.exception(
|
||||||
f"An error occurred updating end user cache: {str(e)}"
|
f"An error occurred updating end user cache: {str(e)}"
|
||||||
|
|
Loading…
Add table
Add a link
Reference in a new issue